Machine Learning-Enhanced Neurocognitive Assessment in Primary Healthcare Settings

Inventor: Ganesh Baliga, PhD and David J. Libon, PhD

Brief Description

The Rowan Assessment of Visual Memory and Attention (RAMA) is an easy-to-administer, non-invasive iPad-based assessment composed of four well-tolerated, non-verbal neuropsychological tests. Employed in research underway at leading hospitals, the RAMA tests generate data that powers a machine learning algorithm for highly accurate cognitive impairment screening.

Problem

With over six million Americans aged 65 and older currently affected by Alzheimer’s Disease and Related Dementia (ADRD), and another 16% suffering from Mild Cognitive Impairment (MCI)—a high-risk precursor to ADRD—the public health crisis is immense. This challenge is further complicated by the fact that many frontline healthcare providers feel unprepared to properly screen, diagnose, and administer new treatments like anti-amyloid, monoclonal antibody therapy. The moment is critical for an easy-to-administer, effective cognitive screening tool that can be seamlessly integrated into a routine physical in a primary care setting.

Solution

RAMA tests are brief, culture-neutral screening tools for early ADRD and administered on an iPad. The tests primarily use a visual format because recent research suggests visual assessment is more sensitive to early emergent ADRD than the prevalent verbal/auditory tests. This visual design also overcomes a key limitation of verbal tests, making the RAMA suite inherently culture-neutral and deployable world-wide. RAMA tests are designed to be highly sensitive to core deficits associated with ADRD, including executive control, working memory, and visual episodic memory. With over 3,400 administrations in research projects at national hospitals validating their effectiveness and scalability, a machine learning algorithm developed using patient data collected at Rowan University School of Osteopathic Medicine memory clinic has further demonstrated RAMA’s potential as a highly accurate screening tool in primary care.

Technology

RAMA is a non-invasive, digitally administered test suite built upon the scholarly foundations of neuropsychology. Delivered via an iPad app, it functions as an effective screen for cognitive impairment while remaining brief, non-invasive and well-tolerated by patients.

Include labeled diagrams, schematics, or data visuals that support the innovation. Examples:

• Device schematic
• Workflow diagram